Zirconium-catalyzed carboalumination of α-olefins and chain growth of aluminum alkyls: Kinetics and mechanism

Article abstract of DOI:10.1021/ja104032w

A mechanism based on Michaelis-Menten kinetics with competitive inhibition is proposed for both the Zr-catalyzed carboalumination of α-olefins and the Zr-catalyzed chain growth of aluminum alkyls from ethylene. AlMe₃ binds to the active catalyst in a rapidly maintained equilibrium to form a Zr/Al heterobimetallic, which inhibits polymerization and transfers chains from Zr to Al. The kinetics of both carboalumination and chain growth have been studied when catalyzed by [(EBI)Zr(μ-Me)₂AlMe₂] [B(C₆F₅)₄]. In accord with the proposed mechanism, both reactions are first-order in [olefin] and [catalyst] and inverse first-order in [AlR₃]. The position of the equilibria between various Zr/Al heterobimetallics and the corresponding zirconium methyl cations has been quantified by use of a Dixon plot, yielding K = 1.1(3) × 10 ^-4 M, 4.7(5) × 10^-4 M, and 7.6(7) × 10 ^-4 M at 40 °C in benzene for the catalyst species [rac-(EBI)Zr(μ-Me)₂AlMe₂][B(C₆F ₅)₄], [Cp₂Zr(μ-Me)₂AlMe ₂][B(C₆F₅)₄], and [Me ₂C(Cp)₂Zr(μ-Me)₂AlMe₂][B(C ₆F₅)₄] respectively. These equilibrium constants are consistent with the solution behavior observed for the [Cp ₂Zr(μ-Me)₂AlMe₂][B(C₆F ₅)₄] system, where all relevant species are observable by ¹H NMR. Alternative mechanisms for the Zr-catalyzed carboalumination of olefins involving singly bridged Zr/Al adducts have been discounted on the basis of kinetics and/or ¹H NMR EXSY experiments.(Figure Presented)

Full text of DOI:10.1021/ja104032w

ARTICLE
pubs.acs.org/JACS
Zirconium-Catalyzed Carboalumination of r-Olefins and Chain
Growth of Aluminum Alkyls: Kinetics and Mechanism
James M. Camara,‡ Robby A. Petros,§ and Jack R. Norton*,†
†Department of Chemistry, Columbia University, New York, New York 10027, United States
‡Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, United States
§Department of Chemistry, University of North Texas, Denton, Texas 76203, United States
S Supporting Information
b
ABSTRACT: A mechanism based on Michaelis-Menten kinetics with com-
petitive inhibition is proposed for both the Zr-catalyzed carboalumination of
R-olefins and the Zr-catalyzed chain growth of aluminum alkyls from ethylene.
AlMe3 binds to the active catalyst in a rapidly maintained equilibrium to form a
Zr/Al heterobimetallic, which inhibits polymerization and transfers chains
from Zr to Al. The kinetics of both carboalumination and chain growth have
been studied when catalyzed by [(EBI)Zr(μ-Me)2AlMe2][B(C6F5)4]. In
accord with the proposed mechanism, both reactions are first-order in
[olefin] and [catalyst] and inverse first-order in [AlR3]. The position of the
equilibria between various Zr/Al heterobimetallics and the corresponding
zirconium methyl cations has been quantified by use of a Dixon plot, yielding K = 1.1(3) ꢀ 10-4 M, 4.7(5) ꢀ 10-4 M, and 7.6(7) ꢀ
10-4 M at 40 °C in benzene for the catalyst species [rac-(EBI)Zr(μ-Me)2AlMe2][B(C6F5)4], [Cp2Zr(μ-Me)2AlMe2][B(C6F5)4],
and [Me2C(Cp)2Zr(μ-Me)2AlMe2][B(C6F5)4] respectively. These equilibrium constants are consistent with the solution behavior
1
observed for the [Cp2Zr(μ-Me)2AlMe2][B(C6F5)4] system, where all relevant species are observable by H NMR. Alternative
mechanisms for the Zr-catalyzed carboalumination of olefins involving singly bridged Zr/Al adducts have been discounted on the
basis of kinetics and/or 1H NMR EXSY experiments.
’ INTRODUCTION
Several groups have explored the catalysis of chain growth in
main group organometallics. In 1991, Samsel showed that chain
growth in organoaluminums could be catalyzed by Zr and Hf
complexes7 and subsequently reported the catalysis of the same
reaction by Cp*2ThCl2.8 Bazan reported the use of Cp*CrMe2-
(PMe3) to catalyze the insertion of ethylene into Al-C bonds,9 and
Gabbaï demonstrated that Cp*Cr(C6F5)R complexes could cata-
lyze the same reaction.10 Chen has used a Cp*ZrCl2/MAO
system11 and Kempe has used an organoyttrium/AlR3 system12
to produce polyethylene with low polydispersity. Gibson has shown
that a bis(imino)pyridine iron complex can catalyze the insertion of
ethylene into Zn-C bonds and has established that Fe/Zn transfer
is fast and reversible throughout such a polymerization (“catalyzed
chain growth”).13 Fast and reversible transfers among an organozinc
compound, a Zr catalyst, and an Hf catalyst have been used to effect
“chain shuttling”, producing block copolymers with different levels
of R-olefin incorporation.14 Recently, Sita and co-workers have
described the use of a Hf catalyst to effect the apparent chain growth
of aluminum alkyl chains; insertion of ethyene into Hf-C bonds is
followed by chain transfer to Al through ZnR2.15
The carbometalation of olefins (eq 1) has found many uses in
synthesis.1 The carboalumination of R-olefins offers particular
potential because the product aluminum alkyl may, in principle,
be used to prepare alcohols, aldehydes, carboxylic acids, and
amidines (Figure 1).2
Dzhemilev and co-workers reported the first zirconium-
catalyzed carboalumination in 1985, using Cp2ZrCl2/MgEt2 to
catalyze the addition of AlEt3 across 1-hexene.3 Negishi reported
a system for the asymmetric carboalumination of R-olefins
using bis(1-neomenthylindene)zirconium dichloride4 as a
catalyst and has applied the methodology to natural product
synthesis.5
Carboalumination reactions play an important role in an
industrial setting. The direct carboalumination of ethylene is
used in the production of fatty alcohols (linear alcohols with
chain lengths between C6 and C22), also known as Ziegler
alcohols. In the Alfol process multiple ethylenes are inserted
into the Al-C bonds of triethylaluminum at high temperature
and pressure; in the Epal process fewer ethylenes are inserted in
each of several short cycles.6 Oxidation and hydrolysis convert
the elongated chains into linear alcohols.
The exchange of alkyl chains between Zr and Al surely occurs via
the formation of a Zr/Al bimetallic complex like 1, first reported by
Bochmann and Lancaster.16 We have previously reported17 the
Received: May 11, 2010
Published: March 23, 2011
r
2011 American Chemical Society
5263
dx.doi.org/10.1021/ja104032w J. Am. Chem. Soc. 2011, 133, 5263–5273
|

Journal of the American Chemical Society
ARTICLE
Figure 1. Potential transformations of aluminum alkyls.
rate constant koff for the exchange of AlMe3 from 1 (and other
such complexes) into (AlMe3)2 in C6D6, presumably by
dissociation as in reaction 2. Both cations in eq 2 are solvated
to some extent by the aromatics (bezene, toluene) in which
carboalumination and chain growth are usually carried out.
Figure 2. First-order exponential fit to the disappearance of allyl-
benzene during methylalumination in the presence of [(EBI)Zr-
(μ-Me)2AlMe2][B(C6F5)4] (eq 4) at 40 °C in C6D6. [(EBI)Zr(μ-
Me)2AlMe2]þ = 0.38 mM, [allylbenzene]o = 2.55 mM, [AlMe3]total
61.2 mM.
=
it may be used to form the known18g heterobimetallic cation
[(EBI)Zr(μ-Me)2AlMe2]þ (3) cleanly from (EBI)ZrMe2 in the
presence of excess (AlMe3)2 as in eq 6.24
Other researchers have observed species similar to 1 and have
suggested that they are involved in chain transfer.18 The X-ray
structure of the related complex [Ti(NtBu)(Me3[9]aneN3)(μ-
Me)2AlMe2]þ has been published, along with the results of
calculations on such methyl-bridged heterobimetallics.19
The mechanism by which Zr catalyzes the carboalumination in
eq 3 has been examined by Shaughnessy and Waymouth.20 Their
study yielded surprising results, with the reaction appearing to be
first-order in substrate (allylbenzene) but second-order in the
zirconium catalyst and of variable order in AlMe3 (-2 at [AlMe3]
between 0.64 and 1.27 M, -0.75 at [AlMe3] > 1.27 M).
However, the concentrations of AlMe3 used were sufficiently
small that [AlMe3] varied during the reactions. Furthermore,
experiments in our laboratories have shown that Me/C6F5
exchange between Al and B is facile under the conditions of
reaction 3 and have identified BMe3 as a product.17 (The
reported synthesis of Al(C6F5)3 involves the reaction of B-
(C6F5)3 and AlMe3.21) Moreover, the use of B(C6F5)3 as an
activator is known to produce a tight ion pair,22 complicating the
reaction kinetics.
We have therefore examined the kinetics and mechanism of
the carboalumination in eq 4 (L2 = EBI, Cp2, and Me2C-
(C5H4)2) and of the AlR3 chain growth in eq 5. To generate the
catalysts we have treated the appropriate dimethyl metallocene
-
with [Ph3C][B(C6F5)4] (B(C6F5)4 is known to be weakly
’ RESULTS AND DISCUSSION
coordinating).23 We have already shown that activation by this
reagent forms Cp2ZrMeþ from Cp2ZrMe2 without Me/C6F5
exchange (the only side product is triphenylethane)17 and that
Kinetics of Zr-Catalyzed Carboalumination (reaction 4).
The dimethyl metallocene precatalyst (e.g., (EBI)ZrMe2) was
5264
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Figure 4. Plot of kobs-1 (s) (reaction 4) measured under pseudo-first-
order conditions vs [AlMe3] in mol/L. [(EBI)Zr(μ-Me)2AlMe2]þ
0.38 mM, [allylbenzene]o = 2.55 mM, T = 40 °C.
=
Figure 3. Plot of kobs (reaction 4) for disappearance of allylbenzene at
40 °C in C6D6 vs [(EBI)Zr(μ-Me)2AlMe2]þ initially added. [allylben-
zene]o = 6.12 mM, [AlMe3]total = 61.2 mM.
The rate law for our carboalumination (reaction 4) is thus first-
order in substrate and [Zr] and inverse first-order in [AlMe3]:
activated with 1 equiv of [Ph3C][B(C6F5)4] in benzene and
converted to the heterobimetallic (e.g., 3) by addition of >2
equiv25 of (AlMe3)2. The disappearance of allylbenzene was
monitored at 40 °C in C6D6 in the presence of 0.15 equiv of
catalyst and a substantial excess (10 equiv) of (AlMe3)2. We saw
no oligomerization. 1H NMR (stack plot in Figure S-1) showed
that the allylbenzene was converted cleanly to the product 2. The
½allylbenzeneꢁ½Zrꢁ
rate ¼ kobs
ð8Þ
½AlMe3ꢁ
Mechanism of Zr-Catalyzed Carboalumination. The me-
chanism in Scheme 1 is consistent with our observed rate law
(eq 8). A reactive zirconium methyl cation B reversibly binds
AlMe3 to form an inactive Zr/Al heterobimetallic A.27 The cation
B coordinates the olefin with rate constant k1, forming C; C
inserts the olefin with rate constant k2, forming a longer-chain
alkyl Zr cation. That cation then reassociates with free AlMe3 to
form a new heterobimetallic A0, resulting in transmetalation and
formation of the product P when A0 dissociates in the direction
shown.
Deriving the rate law for this mechanism is straightforward.
If the formation of A0 and loss of P are faster than chain
growth, the rate of formation of P is given by eq 9. The con-
centration of C (which is not observable) is surely small enough to
justify the steady-state approximation, which gives eq 10.
disappearance of the allylbenzene was cleanly first-order, kobs
=
5.1 ꢀ 10-4 s-1 (Figure 2).
A plot of kobs against the amount of [(EBI)Zr(μ-Me)2AlMe2]þ
initially added gave Figure 3. (The data are in Table S-1.) Reaction
4 is thus first-order in Zr.
Probing the order in AlMe3 proved more challenging. Tri-
methylaluminum exists mostly as the dimer (AlMe3)2 under the
reaction conditions (40 °C in C6D6). The equilibrium constant
KTMA (eq 7) can be estimated from published data for the
dissociation of (AlMe3)2 in mesitylene. Cerny and co-workers
have reported ΔH = 13.1(8) kcal/mol and ΔS = 19(2)
cal/(mol K),26 which imply a ΔG of 7.1 kcal/mol at 40 °C and
3
a KTMA of 1.2 ꢀ 10-5 M at that temperature.
dP
¼ k2½Cꢁ
ð9Þ
dt
d½Pꢁ k2k1½olefinꢁ½Bꢁ
¼
ð10Þ
dt
k-1 þ k2
Noting that the equilibrium in eq 7 is rapidly maintained on
the time scale of the carboalumination reaction,26 and assuming
that the above value of KTMA is valid in benzene, the amount of
monomeric AlMe3 may be calculated from the total amount of
added trimethylaluminum. Measurements of kobs with various
amounts of added trimethylaluminum (but at least a 10-fold
excess over the allylbenzene substrate) gave the results in Table
S-2. Comparison of kobs (for eq 4) with [AlMe3] suggested a rate
law inverse in AlMe3, and a plot of 1/kobs vs [AlMe3] proved
linear (Figure 4).
As the AlMe3-binding equilibrium (eqs 2 and 11) is rapidly
maintained on the carboalumination time scale,17 we can express
[B] as a function of the total zirconium concentration [Zr]T,
[olefin], and [AlMe3] (eq 12).
koff ½AlMe3ꢁ½Bꢁ
½AlMe3ꢁ½Bꢁ
K ¼
¼
w ½Aꢁ ¼
ð11Þ
kon
A
K
5265
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Scheme 1. Proposed Mechanism for the Zr-Catalyzed Carbo-
alumination of R-Olefins (L2 = EBI, Cp2, or Me2C(C5H4)2)
Scheme 2. Michaelis-Menten Kinetics with Competitive
Inhibition
Table 1. Comparison of koff Values (eq 2, 27 °C) with
Dissociation Equilibrium Constants K (40 °C) in C6D6
koff (s-1
)
K = koff/kon (M)
[(EBI)Zr(μ-Me)2AlMe2]þ (3)
[Cp2Zr(μ-Me)2]AlMe2]þ (1)
[Me2C(C5H4)2Zr](μ-Me)2]AlMe2]þ (4)
0.2(1)
1.1(3) ꢀ 10-4
4.7(5) ꢀ 10-4
7.6(7) ꢀ 10-4
0.12(1)
1.8(1)
(analogous to a Dixon plot), as observed in Figure 4. Dividing its
intercept by its slope will give the equilibrium constant K (eq 19).
½ZrꢁT
½Bꢁ ¼
ð12Þ
ð13Þ
½AlMe3ꢁ
k1½olefinꢁ
k-1 þ k2
þ 1 þ
k2½ZrꢁT
K
ꢀ
ꢁ
kobs
¼
ð17Þ
ð18Þ
ð19Þ
k-1 þ k2 ½AlMe3ꢁ
Substitution of eq 12 into eq 10 gives eq 13.
þ 1
k1
K
d½Pꢁ
dt
k2½olefinꢁ½ZrꢁT
ꢀ
ꢁꢀ
ꢁ
¼
1
ðk-1 þ k2Þ
ðk-1 þ k2Þ
k1k2½ZrꢁT
k-1 þ k2
½AlMe3ꢁ
¼
½AlMe3ꢁ þ
þ 1 þ ½olefinꢁ
kobs Kk1k2½ZrꢁT
k1
K
If k2 . k1 and the accumulation of C is negligible, the last term
disappears from the denominator and we obtain a rate law, eq 14,
that is consistent with our kinetic observations.
ðk-1 þ k2Þ
k1k2½ZrꢁT
intercept
¼
¼ K
ðk-1 þ k2Þ
Kk1k2½ZrꢁT
slope
d½Pꢁ
dt
k2½olefinꢁ½ZrꢁT
ꢀ
ꢁꢀ
ꢁ
¼
ð14Þ
k-1 þ k2
½AlMe3ꢁ
The intercept and slope of the plot in Figure 4 imply29 that K for
the EBI heterobimetallic 3 is 1.1(3) ꢀ 10-4 M in benzene at 40 °C.
Similar experiments (the reagent concentrations and the resulting
kobs values are in Table S-2) with the unbridged Cp2 Zr/Al (1) and
the Me2C-bridged Zr/Al (4) give the other K values in Table 1.
Our previously measured17 values of koff at 27 °C (also in
benzene) are given in Table 1 for comparison.30,31 While 4 shows
both a relatively large koff and a relatively large K, the relationship
between koff and K is not in general linear; i.e., kon is not constant
from one zirconocene to another.
Olefin and Catalyst Dependence of the Zr-Catalyzed
Growth of Organoaluminum Chains (eq 5). Determining
the rate law for such a reaction is challenging. Use of ethylene
requires a pressure vessel. Zirconium methyl cations are highly
active catalysts, and chain growth may become limited by mass
transport; rate measurements will not be informative unless the
reaction is slower than the rate at which the ethylene dissolves.
The growth reaction is exothermic making temperature control
difficult. Finally, the growth of the chains during the reaction
leads to an increase in the viscosity of the solution, which affects
the rate of further chain growth.
The apparatus in Figure 5 enabled us to monitor ethylene
uptake as a function of time (details in Supporting Information).
A flow meter calibrated for ethylene was connected to a mass
totalizer, which recorded the total amount of ethylene that had
been admitted to the system as a function of time. The ethylene
passed through copper cooling coils into a multiport pressure
vessel with a thermocouple, pressure gauge, pressure relief valve,
injection port, and mechanical stirrer. The vessel was immersed
as completely as possible in a 0 °C cold bath, which required that
þ 1
k1
K
Scheme 1 is analogous to the competitive inhibition of the
Michaelis-Menten situation in Scheme 2, which gives the
familiar rate law28 in eq 15; a negligible accumulation of E•S
reduces that rate law to eq 16.
d½Pꢁ
dt
k2½Sꢁ½EꢁT
ꢀ
ꢁꢀ
ꢁ
¼
ð15Þ
k-1 þ k2
k1
½Iꢁ
þ 1 þ ½Sꢁ
K1
d½Pꢁ
dt
k2½Sꢁ½EꢁT
ꢀ
ꢁꢀ
ꢁ
¼
ð16Þ
k-1 þ k2
½Iꢁ
þ 1
k1
K1
The rate laws in eqs 14 and 16 are the same. In eq 14 the
zirconium alkyl cation B is the “enzyme”, the Zr/Al heterobime-
tallic A is the “enzyme-inhibitor complex”, and the olefin is the
“substrate”, S. The only significant difference between Schemes 1
and 2 is the fact that AlR3 is both an inhibitor and a reactant in
Scheme 1. This does not affect the rate law because AlR3 serves as
a reactant only after the rate-determining step.
Determination of Equilibrium Constants K for the Disso-
ciation of AlMe3 from Zr/Al Heterobimetallic Cations. Plots
like those in Figure 4 allow for the determination of the
equilibrium constant K (eq 2). From eq 14, kobs for the
disappearance of allylbenzene will be given by eq 17, and
inversion of eq 17 gives the expression in eq 18. From eq 18, it
is clear that a plot of 1/kobs vs [AlMe3] should be a straight line
5266
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Figure 5. Diagram of the reactor used for measuring chain growth kinetics.
the benzene used as solvent in our carboalumination studies be
replaced with toluene. The reaction proved sufficiently exother-
mic that its temperature rose slightly above that of the bath, being
about 1.5 ( 0.5 °C during data collection.
The precatalyst (EBI)ZrMe2 was again activated with 1 equiv
of [Ph3C][B(C6F5)4] and converted to the heterobimetallic 2 by
addition of >2 equiv of (AlMe3)2. Admitting ethylene (20 psig)
to the evacuated reaction gave uptake plots like that in Figure 6.
Initial fast absorption as the reaction vessel filled and ethylene
dissolved in the toluene gave way to a modest but constant slope
after the solution had become saturated with ethylene. This
slope, due to chain growth, gave the rate of the reaction. It was
unaffected by a change in the stirring speed, verifying that the rate
was not limited by mass transport.
The concentrations of ethylene at various pressures were
determined from ethylene uptake measurements on pure toluene
(Table 2). The results were consistent with those reported by
other researchers.32
Our kinetics experiments at ethylene pressures >40 psig showed
chain growth to be so fast that it became limited by mass transport. At
ethylene pressures <20 psig the reaction was so slow that the mass
totalizer did not give reliable readings. Within that pressure range (20
to 40 psig), however, the rate of chain growth appeared to be linear
with [C2H4] (Table S-3 and Figure S-2). At 20 psig ethylene,
variation of [Zr]tot showed the rate of chain growth to be linear in
catalyst (Table S-4 and Figure S-3).
Figure 6. Typical ethylene uptake curve obtained from the reactor for a
catalyzed chain growth reaction (T = 0 °C, Pethylene = 20 psig, [(EBI)Zr-
(μ-Me)2AlMe2]þ = 4.39 μM, [AlMe3] = 0.439 M).
Products from Zr-Catalyzed Growth of Organoaluminum
Chains (eq 5) and Estimation of Monomeric AlR3. Hydrolysis
of the organoaluminum products from a typical experiment
(20psig ethylene, 4.39 μM catalyst, 439 mM AlMe3, 6 min reaction
time) resulted in considerable gas evolution, implying that many
Al-Me bonds remained unchanged after ethylene uptake.
Considerable polymer (∼150 mg) precipitated, which GPC
showed to have an Mw of 92000, an Mn of 29000, and a PDI of
3.2 (Figure S-4). 1H NMR showed the polymer to contain almost
no unsaturation (Figure S-5). If the organoaluminum products
were exposed to O2 before hydrolysis, no gas was evolved.
5267
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Table 2. Solubility of Ethylene in Toluene at 0 °C Measured
at Various Pressures
Pressure (psig)
[Ethylene] (M)
20
30
40
0.44
0.59
0.79
Table 3. Equilibrium Constants for Dissociation of (AlR3)2
for Various Trialkylaluminums at 50 °C in Benzene33
AlR3
KAl
AlMe3
7.4 ꢀ 10-7
1.4 ꢀ 10-4
2.5 ꢀ 10-3
5.0 ꢀ 10-3
7.1 ꢀ 10-3
8.2 ꢀ 10-3
9.1 ꢀ 10-3
6.0
AlEt3
Al(n-Pr)3
Al(n-Bu)3
Al(n-Hep)3
Al(n-Oct)3
Al(n-Dec)3
Al(i-Bu)3
Figure 7. Dependence of Zr-catalyzed chain growth on [AlR3] (eq 5).
[AlR3] was calculated from the equilibrium constant for dissociation of
trimethylaluminum26 at 0 °C in mesitylene (Pethylene = 20 psig, [(EBI)-
Zr(μ-Me)2AlMe2]þ = 4.39 μM).
The distribution of chain lengths implied by these results
made it difficult to quantify the monomeric AlR3 that was
inhibiting ethylene uptake. The rates were typically measured
>200 s after the reaction was initiated, by which time there had
been significant ethylene uptake. Some idea of the effect of
chain length on association can be obtained from the reported
equilibrium constants KAl (for dissociation of AlR3 dimers) in
Table 3. The values of KAl are not significantly affected by
chain length past butyl, although they are affected by chain
branching.
In order to determine how many Al-C bonds inserted
ethylene during a typical reaction and what portion of the
ethylene became polymer, we performed the following experi-
ment. A chain growth reaction was conducted under typical
reaction conditions (20 psig ethylene, 4.39 μM catalyst, 439 mM
AlMe3), with the ethylene flow restricted to no more than 300
SCC/min by the flow controller; under these conditions we
could measure accurately the consumption of ethylene through-
out the reaction.34 The reaction time was extended to 8 min to
ensure that the solution had become saturated before rate
measurement; the rate of ethylene uptake after saturation was
the same as that observed in previous experiments with the same
Zr, Al, and C2H4 concentrations. A total of 108 SCC of ethylene
(∼123 mg, or 4.4 mmol) was consumed during this time (Figure
S-9).
After quenching and hydrolysis, polymer (106 mg) again
precipitated, containing ∼85% of the ethylene consumed. The
organic layer was separated, neutralized, dried, and analyzed by
1H NMR and MS. Its 1H NMR showed the presence of
oligomeric material, as evidenced by the ratio of methylene H’s
to methyl H’s (≈ 13:1). The polymer was analyzed by GPC and
found to have Mn = 67 500, Mw = 172 500, and PDI = 2.56.
Analysis of the polymer by 1H NMR showed almost no
unsaturation, implying that chain transfer to aluminum was
the principal termination mechanism. In view of the length of
the polymer chains, and the small number of ethylenes con-
sumed per Al (4.4 mmol ethylene per 43.9 mmol of Al), it was
apparent that chain growth had occurred in only a small portion of
the AlMe3.
Chain growth experiments were conducted with Al(oct)3 in place
of AlMe3. In this system, the shortest chain produced upon
hydrolysis would be nonvolatile C8. A plot of total ethylene uptake
vs time implied a rate of 0.18 SCC/s, only slightly greater than the
ratesdeterminedforthesameexperimentwithAlMe3. Hydrolysisof
the products from a quenched reaction under typical conditions (20
psig ethylene, 439 mM Al, 4.39 μM catalyst) produced only polymer
and octane. The absence of any detectable short oligomers suggests
that Al(oct)3 is a less effective inhibitor of catalysis than AlMe3—not
surprising in view of the decrease in KAl from trioctylaluminum to
trimethylaluminum (Table 3). Importantly, these data verify that
polymer is the principal product from chain growth.
In chain growth experiments beginning with trimethylalu-
minum, unreacted AlMe3 and its dimer remain the predominant
organoaluminum species throughout the experiment. Thus, the
concentration of inhibiting monomeric AlR3 in an experiment
can be estimated from KAl for AlMe3 and the amount of
trimethylaluminum initially added. The data (Table S-3)
show that the rate of chain growth declines with increasing
[AlMe3], and a plot of 1/rate vs [AlR3] is linear (Figure 7). To
ensure that this observation is not coincidental with our
choice of KAl, we have also estimated the concentration of
monomeric AlR3 from KAl for trioctylaluminum (Figure
S-10);33d the plot of 1/rate vs [AlR3] is still linear. The
growth of our organoaluminum chains (reaction 5) thus
obeys a rate law like the one (eq 8) that we established for
carboalumination (reaction 4).
The agreement of the observed rate law for chain growth with
that of carboalumination suggests that chain growth occurs by a
mechanism (Scheme 3) like the one in Scheme 1 for carboalu-
mination. A kinetic analysis like that for Scheme 1 gives the rate
law in eq 20 from Scheme 3. If the olefin complex C does not
accumulate, eq 20 reduces to eq 21 (just as eq 13 reduced to
eq 14 for Scheme 1). Inversion of eq 21 gives eq 22, which shows
5268
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Scheme 3. Zirconium-Catalyzed Chain Growth of Alumi-
num Alkylsa
Scheme 4. Alternative Mechanism for Zr-Catalyzed Carbo-
alumination of Olefins Involving a Singly Bridged,
Coordinatively Unsaturated and Catalytically Active Zr/Al
Heterobimetallic D (L2 = Cp2, SBI, EBI, and similar)
a At short reaction times R = Me, at longer reaction times R = P.
that a plot of 1/rate vs [AlR3] should be linear.
Scheme 5. Exchange of Bridging and Terminal Methyl
Groups by Rotation around the Al-C Bond of D
d½Pꢁ
dt
k2½olefinꢁ½ZrꢁT
ꢀ
ꢁꢀ
ꢁ
¼
ð20Þ
ð21Þ
k-1 þ k2
½AlR3ꢁ
þ 1 þ ½olefinꢁ
k1
K
ꢀ
ꢁꢀ
ꢁ
k-1 þ k2
½AlR3ꢁ
if ½olefinꢁ,
þ 1 then
K
k1
d½Pꢁ
k2½olefinꢁ½ZrꢁT
Integration of the spectrum with 1.0 equiv of AlMe3 showed
that the residual [Cp2ZrMe][B(C6F5)4] was ∼0.43 mM, in good
agreement with the estimated value of 0.5(1) mM calculated
from our measurement of K for 1 (4.7(5) ꢀ 10-4 M).
Alternatives to Schemes 1 and 3. The alkyl bridges in
Schemes 1 and 3 break in a pairwise fashion; i.e., the two alkyl
bridges in A open synchronously to the mononuclear B and
AlMe3. It is possible that one alkyl bridge opens more rapidly
than the other, giving a singly bridged intermediate like D in
Scheme 4. (For simplicity we write the bridges as methyl groups.)
Such an intermediate could coordinate the substrate olefin (rate
constant k3) without further dissociation.
ꢀ
ꢁꢀ
ꢁ
¼
k-1 þ k2
½AlR3ꢁ
dt
þ 1
k1
K
ꢀ
ꢁꢀ
ꢁ
k-1 þ k2
½AlR3ꢁ
þ 1
1
d½Pꢁ
dt
k1
K
¼
k2½olefinꢁ½ZrꢁT
ðk-1 þ k2Þ
ðk-1 þ k2Þ
k1k2½ZrꢁT
¼
½AlMe3ꢁ þ
ð22Þ
Kk1k2½ZrꢁT
If coordination is rate determining, the rate law derived for this
mechanism is given in eq 23. The reciprocal of the corresponding
observed pseudo-first-order rate constant is given by eq 24.
(Derivations of these equations are given in the Supporting
Information.) It is clear that eq 24 is not linear in [AlMe3], so
Scheme 4 is inconsistent with our kinetic observations (or at least
not supported by them).
It is, however, not practical to determine the equilibrium
constant K (for the dissociation of 3) from the intercept and
slope of the plot in Figure 7. There is not a simple dissociation
equilibrium, but rather many dissociation equilibria involving
various AlR3 present. The concentrations of monomeric AlR3 are
only estimates, and the plot in Figure 7 (while linear enough to be
consistent with eq 22 and Scheme 3) does not determine the
intercept with precision.
dP
dt
k3½olefinꢁ½AlMe3ꢁ½ZrꢁT
k4½olefinꢁ½ZrꢁT
ꢀ
ꢁ
-
¼
þ
ð23Þ
½AlMe3ꢁ ½AlMe3ꢁ
½AlMe3ꢁ ½AlMe3ꢁ
Direct Determination of the Dissociation Equilibrium
Constants K for the Cp Heterobimetallic 1. In principle, the
values of K determined by eqs 18 and 19 from carboalumination data
can be checked by direct examination of the appropriate dissociation
equilibria (eq 2). However, the complexity of their 1H NMR spectra
makes it hard to quantify the dissociation of the EBI heterobimetallic
3 or the Me2C-bridged heterobimetallic 4.
K2 1 þ
þ
1 þ
þ
K2
K1K2
K2
K1K2
ꢀ
ꢁ
½AlMe3ꢁ ½AlMe3ꢁ
K2 1 þ
þ
1
K2
K1K2
¼
ð24Þ
kobs
k3½AlMe3ꢁ½ZrꢁT þ K2k4½ZrꢁT
In contrast the resonances of [Cp2ZrMe][B(C6F5)4]35 are
easily distinguished from the resonances of the Zr/Al hetero-
bimetallic 1. When [Cp2ZrMe][B(C6F5)4] was generated from
If k3 is negligible, Scheme 4 becomes kinetically equivalent to
Schemes 1 and 3. If k4 is negligible we obtain a derived rate law
with kobs given by eq 25, which cannot be linear in [AlMe3] and is
inconsistent with the observed kinetics.
1
Cp2ZrMe2 and [Ph3C][B(C6F5)4], its H NMR spectrum in
benzene at 40 °C (Figure S-6) showed clean conversion to
[Cp2ZrMe][B(C6F5)4]. The addition of exactly 1.0 equiv of
trimethylaluminum did not convert all of the [Cp2ZrMe][B(C6-
F5)4] to 1 (Figure S-7), although the conversion became quanti-
tative after the addition of excess AlMe3.36 Stack plots of the 1H
NMR spectra taken as AlMe3 was added are shown in Figure S-8.
1
K2
1
1
¼
þ
þ
ð25Þ
kobs k3½AlMe3ꢁ½ZrꢁT k3½ZrꢁT K1k3½ZrꢁT
An even more convincing argument against the presence of a
significant amount of a singly bridged Zr/Al species such as D is
5269
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Scheme 6. Fast Insertion Relative to Coordination of Olefin
to Zirconium Alkyl Cation
Scheme 7. Rapid Unfavorable Binding of Olefin Followed by
Rate-Determining Insertion
’ CONCLUSIONS
Figure 8. 1H NMR EXSY spectrum of [Cp2Zr(μ-Me)2AlMe2][B-
(C6F5)4] (1) in benzene at 40 °C with a mixing time of 300 ms. AlMe3
(δ = -0.38) is seen to exchange with the bridging methyl groups (δ =
-0.46) and the terminal methyl groups (δ = -0.77).
The kinetics we have observed for carboalumination (reaction
4) strongly support the mechanism in Scheme 1, and our data on
Zr-catalyzed chain growth (reaction 5) are consistent with the
analogous mechanism in Scheme 3. The coordination of AlR3 to
the catalytically active organozirconium cation (B) is a preequilib-
rium that lies toward the (μ-R)2 heterobimetallic (A) under
reaction conditions but is rapidly maintained on the time scale of
the overall reaction. However, actual transfer of the growing
chains appears to be somewhat slow, which may reflect a strong
preference for bridging methyls over longer chain bridging alkyl
groups.
The fact that no [olefin] term is observed in the denominator
of either rate law (i.e., that eq 13 reduces to eq 14, and eq 20 to
eq 21) implies that the concentration of the olefin complex C is
negligible compared to that of the heterobimetallic A or the
cation B. Two explanations are possible: either insertion is fast
relative to the coordination of the olefin (Scheme 6) or the olefin
is bound in a rapid but unfavorable equilibrium before rate-
determining insertion (Scheme 7). There is experimental sup-
port for Scheme 7 in the case of incoming alkynes39 and
1-alkenes.40 With incoming ethylene, chain growth is too fast
to permit rate measurements and is usually diffusion controlled.
The relative efficiency of various Zr complexes as catalysts may
be a function of k1 (if Scheme 6 is correct) or k1k2/k-1 (if
Scheme 7 is correct). However, a major factor—likely to be
dominant in many cases—will be the size of the equilibrium
constant K for the dissociation of the corresponding Zr/Al
heterobimetallic.
offered by the 1H NMR of Zr/Al heterobimetallics. Formation of
D, and rotation around its Al-C bond, would result in fast direct
exchange of the terminal and bridging methyl groups (Scheme 5).
Direct exchange of this type should result in readily observable
bridge/terminal cross peaks in the 1H NMR EXSY spectrum of
solutions of a heterobimetallic A. However, no such exchange is
seen for 1 in Figure 8. (Variation of mixing time makes no
difference.) The bridging methyls and the terminal methyls of 1
do, as we would expect from the operation of the equilibrium in
eq 2, both exchange with free trimethylaluminum. Trimethylalu-
minum itself gives similar results: exchange between the bridging
and the terminal methyl groups of the dimer (AlMe3)2 takes place
principally by dissociation into the AlMe3 monomer.26
The fact that there is no observable direct exchange between the
bridging and terminal methyl groups of A argues against noncom-
petitive inhibition,37 in which the substrate binds to D rather than
to the Zr methyl cation B and forms product at an attenuated rate.
While such mechanisms can in principle be tested kinetically, it is
not practical in our system38 to achieve the required range of
concentrations of the AlMe3 monomer.
Catalyzed Chain Growth and Catalytic Chain Growth.
Gibson and co-workers have distinguished “catalyzed chain
growth”, with fast and reversible exchange of the polymer chain
between a transition metal and Al, from “catalytic chain growth”,
in which the transition metal catalyzes olefin insertion and chain
growth without implying anything about the rate of chain
exchange between the transition metal and Al.13a Catalyzed chain
growth produces a Poisson distribution of molecular weights;
modeling studies13a show that the rate of Zr/Al transfer must be
at least 100 times faster than the rate of propagation in order to
achieve such a distribution.
Our kinetics confirm that AlR3 monomers bind to the Zr
cations in “competitive inhibition” preequilibria but reveal
nothing about how rapidly chains are exchanged between Zr
and Al. However, most of the product in our experiments is
polyethylene of high polydispersity, implying that insertion into
Zr-C is much faster than chain transfer from Zr to Al. Thus, our
chain growths are not “catalyzed chain growth reactions” as
defined by Gibson and co-workers.13a
’ EXPERIMENTAL SECTION
General Considerations. An inert atmosphere and Schlenk line
methods were used to handle air and moisture sensitive compounds.
Benzene was distilled from sodium/benzophenone under N2 prior to
use. Diethyl ether and toluene were purified using a standard Grubbs-
type solvent system. Hexanes were first deolefinated by stirring with
H2SO4, then dried over sodium, distilled, and stored under argon. Stock
solutions were made inside a glovebox and stored in a -20 °C freezer
when not in use.
Synthesis of (EBI)ZrMe2. MeMgBr (1.7 mL, 3 M in hexanes) was
added to ∼50 mL of diethylether in a Schlenk flask under nitrogen.
Dioxane (0.5 mL) was then added, generating MgMe2 and MgBr2
(immediate white precipitate). After stirring for ∼1 h, the solids were
allowed to settle and the solution was transferred by filter cannula to
5270
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
-
another Schlenk flask containing rac-(EBI)ZrCl2 (0.75 g, 1.56 mmol)
slurried in ∼50 mL of diethyl ether at -78 °C. The reaction was allowed
to warm to room temperature overnight. The solvent was then removed,
leaving a yellow solid. This solid was dissolved in benzene, resulting in a
yellow solution and a white solid (MgCl2). The solution was then
filtered and the solvent removed under vacuum, leaving 620 mg (91%
yield) of a pure yellow solid. The product may be recrystallized from
Inhibition Kinetics of [Cp2Zr(μ-Me)2AlMe2þ][B(C6F5)4
]
and [Me2C(Cp)2Zr(μ-Me)2AlMe2þ][B(C6F5)4-]. The procedure
for dependence on AlMe3 was followed as above except that 12.6 mg of
Cp2ZrMe2 or 14.6 mg of Me2C(Cp)2ZrMe2 were substituted for 18.9
mg of rac-(EBI)ZrMe2. For these catalysts it was necessary to add ∼5
equiv of AlMe3 to the activated species [L2ZrMeþ][B(C6F5)4-] to
quantitatively form the corresponding Zr/Al heterobimetallic.
Stoichiometric Reaction of [Cp2Zr(μ-Me)2AlMe2þ][B-
(C6F5)4-] with AlMe3. A stock solution of activated catalyst was
prepared in the glovebox by adding Cp2ZrMe2 (12.6 mg, 0.04 mmol) to
[Ph3C][B(C6F5)4] (46.1 mg, 0.04 mmol) and ∼10 mL of C6D6 in an
amber vial wrapped in aluminum foil. The solution was allowed to sit
overnight. The solution was then filtered through glass wool into
another amber vial. Into a J. Young tube was transferred 0.5 mL of the
solution and 0.100 mL of 9.388 mM toluene in C6D6 as an internal
standard. A 1H NMR spectrum was acquired at 40 °C with a delay time
of 12 s to ensure accurate integrations of all peaks (the T1 of the toluene
methyl resonance is 2.3 s under these conditions). The concentration of
the zirconium methyl cation was calculated against the toluene standard.
Meanwhile, 5 μL of neat AlMe3 was diluted to 5 mL in C6D6 to yield a
0.0104 M solution. The appropriate quantity of this solution was then
added to the J-Young tube containing the zirconium methyl cation. After
allowing the tube to equilibrate to the 40 °C probe temperature, another
1H NMR spectrum was acquired. Representative spectra can be found in
Figures S6-S8.
toluene/hexanes. 1H NMR (500 MHz, C6D6): δ 7.31 (d, C6H, JH-H
=
8.5 Hz, 2H), 7.06 (m, C6H, 4H), 6.89 (m, C6H, 2H), 6.41 (d, C5H,
JH-H = 3.3 Hz, 2H), 5.65 (d, C5H, JH-H = 3.3 Hz, 2H), 2.83-2.66 (m,
CH2CH2, 4H), -0.97 (s, ZrMe2, 6H).
Synthesis of Cp2ZrMe2.41 Cp2ZrCl2 (3.0 g, 10.26 mmol) was
added to a Schlenk flask with a stir bar. Dry Et2O (150 mL) was added by
cannula to the flask. The flask was cooled to -78 °C, and 19.25 mL of
1.6 M LiMe/MgBr (30.78 mmol) were added slowly by syringe with
stirring. The reaction was allowed to warm to room temperature over the
course of 3 h. Chlorotrimethylsilane (1.3 mL, 10.26 mmol) was then
added by syringe, and the reaction was stirred for 1 h. The volatiles were
removed under vacuum. The product was extracted away from the
lithium salts by the addition of 50 mL of benzene followed by cannula
filtration. The benzene was removed under vacuum, leaving 2.21 g of
white solid (86%). 1H NMR (500 MHz, C6D6): δ 5.72 ppm (s, C5H5,
10H), -0.129 ppm (s, CH3, 6H).
Synthesis of Me2C(Cp)2ZrMe2. The same procedure for the
preparation of Cp2ZrMe2 was followed with the substitution of Me2C-
(Cp)2ZrCl2 for Cp2ZrCl2. The reaction was scaled down to 300 mg
(0.9 mmol) of Me2C(Cp)2ZrCl2 starting material and yielded 190 mg
General Procedure for Kinetic Studies of Chain Growth. A
stock solution of activated catalyst was prepared in the glovebox as
described for carboalumination.
of product (72%). 1H NMR (500 MHz, C6D6): δ 6.33 ppm (d, JH-H
=
The flow controller and flow meter were turned on at least 30 min
prior to use to allow the instruments to warm up and equilibrate. Inside a
glovebox, the desired amounts of catalyst and AlMe3 were added to
enough toluene to make a 100 mL solution. The solution was added to
the pressure reaction vessel, and the vessel was closed inside the
glovebox, removed, and placed in the recirculating cold bath to cool.
During this time it was connected to the manifold, thermocouple meter,
mechanical stirrer, and ethylene feed. The mechanical stirrer was set to a
75% maximum stir rate. The reactor was evacuated and backfilled with
argon twice and then evacuated a final time before the reaction began.
Before opening the ethylene valve, the desired pressure was set on the
regulator and the flow controller was fully open for initial pressurization.
The reaction was started by opening the ethylene feed. Caution: Always
shield pressurized vessels! When the flow meter showed that the rate of
ethylene uptake was below 300 SCC (the maximum of accurate
measurement), the flow controller was switched from fully open to
the “controlled” position, allowing accurate measurement of flow. The
reaction was allowed to proceed for 5 min, and then the stir rate was
increased to maximum for an additional minute. Typically, the reactor
was completely pressurized after 1.5 min of total elapsed time. Reactions
were quenched by injection of 3.0 mL of triethylamine.
Pressure Dependence. Reactions were carried out with 4.2 mL
of AlMe3 (0.439 M in Al) and 4.39 ꢀ 10-6 M [rac-(EBI)Zr(μ-Me)2-
AlMe2þ][B(C6F5)4-]. The pressures tested were 20, 30, and 40 psig.
Catalyst Dependence. Reactions were carried out with 4.2 mL of
AlMe3 (0.439 M in Al) and 20 psig of ethylene pressure. The concentra-
tions of [rac-(EBI)Zr(μ-Me)2AlMe2þ][B(C6F5)4-] catalyst tested were
8.78 ꢀ 10-6, 4.39 ꢀ 10-6, 2.195 ꢀ 10-6, and 1.098 ꢀ 10-6 M.
AlR3 Dependence. Reactions were carried out under 20 psig of
ethylene pressure with 4.39 ꢀ 10-6 M [rac-(EBI)Zr(μ-Me)2AlMe2þ]-
[B(C6F5)4-]. Enough trimethylaluminum was added to produce solu-
tions 0.439, 0.220, 0.110, 0.055, and 0.329 M in Al.
3.0 Hz, C5H, 4H), 5.11 ppm (d, JH-H = 3.0 Hz, C5H, 4H), 1.12 ppm
(s, CCH3, 6H), 0.03 ppm (s, ZrCH3, 6H).
General Procedure for Kinetic Studies of Carboalumina-
tion. A stock solution of activated catalyst was prepared in the glovebox
by adding (EBI)ZrMe2 (18.9 mg, 0.05 mmol) to [Ph3C][B(C6F5)4]
(46.1 mg, 0.05 mmol) and 10 mL of C6D6 in an amber vial wrapped in
aluminum foil. The solution was allowed to sit overnight. Trimethyl-
aluminum (10 μL, 0.11 mmol) was added, and the solution was allowed
to react for 1 h. The solution was then filtered through glass wool into a
1
new vial, and the concentration catalyst was determined by H NMR
against a toluene internal standard. Typical concentrations of catalyst
stock solutions were 1-2 mM. A C6D6 stock solution that was 61.2 mM
in allylbenzene and 61.2 mM in toluene and a 612 mM (in Al) stock
solution of AlMe3 in C6D6 were prepared.
For each run, the probe temperature was set to 40 °C and calibrated
with an ethylene glycol standard. Each set of conditions was repeated for a
total of three consistent trials. Samples were prepared at the same time and
1
frozen until use. Reactions were monitored by H NMR at 500 MHz.
Threescans were taken every 120 s with a 3 s acquisition time and 1 s delay,
making the time between points 132 s. Integration of the allylbenzene
vinylic peaks (5.05-5.10 ppm) versus the toluene methyl group as a
standard was used for the fitting procedure. The data were imported into
Kaleidagraph and fit to a two-parameter first-order decaying exponential.
Dependence on Catalyst. Samples were all 500 μL total volume.
To a J. Young tube containing 50 μL of AlMe3 solution was added the desired
amount of catalyst solution. Enough C6D6 to give a total volume of 450 μLwas
then added, and the tube was frozen in a -20 °C freezer. Once frozen, 50 μL
of allylbenzene solution were added. The tube was resealed and then frozen
until use. Catalyst concentrations tested were 0.275, 0.455, and 0.691 mM.
Dependence on AlMe3. Samples were prepared exactly as above
except for the following. The total reaction volume was 600 μL, AlMe3
was varied, 25 μL of allylbenzene solution were used, and the catalyst
loading was 15% with respect to allylbenzene for each run. Enough
trimethylaluminum was added to produce solutions 18.4, 36.7, 73.4,
91.8, and 275 mM in Al—which were respectively 0.46, 0.65, 0.93, 1.04,
and 1.8 mM in monomeric AlMe3.
Hydrolysis (and Oxidation) of Aluminum Alkyls from
Chain Growth Experiments. The reaction mixture was quenched
by injection of 3 mL of triethylamine into the reactor. The solution was
then poured slowly into 500 mL of acidified (with HCl) methanol.
5271
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
Additional HCl was added until all aluminum salts were dissolved,
leaving the precipitated polymer. The polymer was collected by suction
filtration and dried for 24 h at 60 °C in a vacuum oven. The resulting
polymer was then analyzed by GPC and 1H and 13C NMR (see spectra
in the Supporting Information). For samples that required oxidation, dry
O2 was bubbled through the reaction solution for 8 h after quenching.
Precipitation of the resulting products was performed as above.
’ ACKNOWLEDGMENT
We are grateful to Bryan Coughlin for advice on the design of
the equipment in Figure 5, to Exxon for the molecular weight
determinations, to Boulder Scientific for compounds, and to the
National Science Foundation (CHE 0749537) for support. We
thank Condea Vista (now SASOL) for support of the initial
phase of this investigation. We also thank A. K. Rappe, Frank Rix,
and Clark Landis for helpful discussions.
Controlled Flow Chain Growth Reaction. A chain growth
reaction (20 psig ethylene, 4.39 ꢀ 10-6 M [rac-(EBI)Zr(μ-Me)2-
AlMe2þ][B(C6F5)4-], 0.439 M in Al) was performed with the follow-
ing changes to the general procedure. First, the mass flow controller was
allowed to restrict the flow of ethylene to no more than 300 SCC/min
over the entire course of the reaction. Second, the reaction time was
extended to 8 min to ensure saturation before rate measurement. Third,
hydrolysis of the quenched product mixture was carried out by pouring
it into 300 mL of 20% HCl (aq) with vigorous stirring. Polymeric
product (106 mg) immediately precipitated and was collected by
filtering the biphasic solution. The organic layer of the filtrate was then
separated, and the aqueous layer washed with toluene (2 ꢀ 25 mL). The
combined organic layers were washed with 1.0 M NaHCO3 (aq) and
dried over Na2SO4. The rate of the reaction was determined to be 0.143
SCC/min, in excellent agreement with the rates determined from other
experiments at these concentrations (see Supporting Information). 1H
NMR analysis of the toluene solution of the product indicated the
’ REFERENCES
(1) Fallis, A. G.; Forgione, P. Tetrahedron 2001, 57, 5899–5913.
(2) (a) Mason, M. R.; Song, B.; Kirschbaum, K. J. Am. Chem. Soc. 2004,
126, 11812–11813. (b) Lehmkuhl, H. Angew. Chem. 1964, 76, 817. (c)
Rowley, C. N.; DiLabio, G. A.; Barry, S. T. Inorg. Chem. 2005, 44, 1983–
1991. (d) Chang, C. C.; Hsiung, C. S.; Su, H. L.; Srinivas, B.; Chiang, M. Y.;
Lee, G. H.; Wang, Y. Organometallics 1998, 17, 1595–1601.
(3) Dzhemilev, U. M.; Ibragimov, A. G.; Vostrikova, O. S.; Tolstikov,
G. A. Bull. Acad. Sci. USSR, Div. Chem. Sci. 1985, 34, 196–197.
(4) (a) Erker, G. Pure Appl. Chem. 1992, 64, 393–401. (b) Erker, G.;
Aulbach, M.; Knickmeier, M.; Wingbermuhle, D.; Kruger, C.; Nolte, M.;
Werner, S. J. Am. Chem. Soc. 1993, 115, 4590–4601.
(5) (a) Negishi, E.-I.; Kondakov, D. Y. Chem. Soc. Rev. 1996, 25, 417–
426. (b) Negishi, E. Pure Appl. Chem. 2001, 73, 239–242. (c) Negishi, E.-I.;
Huo, S. Q. Pure Appl. Chem. 2002, 74, 151–157. (d) Negishi, E., Tan, Z.
Topics in Organometallic Chemistry; Springer: Berlin, Heidelberg, 2004; Vol.
8, pp 139-176. (e) Negishi, E.-I. Dalton Trans. 2005, 827–848.
(6) Noweck, K. Ullmann’s Encyclopediaof Industrial Chemistry, 7th
ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, 2006.
(7) Samsel, E. G. Catalyzed Chain Growth Process. Patent
US5210338.
(8) Samsel, E. G.; Eisenberg, D. C. Chain growth reactions of R-
olefin(s) - by chain growth on an aluminium alkyl using a activated
actinide metallocene, useful for production of linear olefin(s) and alcohol
(s). Patent US5276220.
(9) Bazan, G. C.; Rogers, J. S.; Fang, C. C. Organometallics 2001, 20,
2059–2064.
(10) (a) Ganesan, M.; Gabbai, F. P. Organometallics 2004, 23, 4608–
4613. (b) Mani, G.; Gabbai, F. P. Angew. Chem., Int. Ed. 2004, 43, 2263–
2266. (c) Ganesan, M.; P. Gabbai, F. J. Organomet. Chem. 2005, 690,
5145–5149.
(11) Rouholahnejad, F.; Mathis, D.; Chen, P. Organometallics 2010,
29, 294–302.
(12) (a) Kretschmer, W. P.; Meetsma, A.; Hessen, B.; Schmalz, T.;
Qayyum, S.; Kempe, R. Chem.—Eur. J. 2006, 12, 8969–8978. (b)
Kretschmer, W. P.; Bauer, T.; Hessen, B.; Kempe, R. Dalton Trans.
2010, 39, 6847–6852.
(13) (a) Britovsek, G. J. P.; Cohen, S. A.; Gibson, V. C.; van Meurs,
M. J. Am. Chem. Soc. 2004, 126, 10701–10712. (b) van Meurs, M.;
Britovsek, G. J. P.; Gibson, V. C.; Cohen, S. A. J. Am. Chem. Soc. 2005,
127, 9913–9923.
(14) (a) Arriola, D. J.; Carnahan, E. M.; Hustad, P. D.; Kuhlman, R.
L.; Wenzel, T. T. Science 2006, 312, 714–719. (b) Kuhlman, R. L.;
Wenzel, T. T. Macromolecules 2008, 41, 4090–4094. (c) Hustad, P. D.;
Kuhlman, R. L.; Carnahan, E. M.; Wenzel, T. T.; Arriola, D. J.
Macromolecules 2008, 41, 4081–4089. (d) Hustad, P. D.; Kuhlman, R.
L.; Arriola, D. J.; Carnahan, E. M.; Wenzel, T. T. Macromolecules 2007,
40, 7061–7064. (e) Wenzel, T. T.; Arriola, D. J.; Carnahan, E. M.;
Hustad, P. D.; Kuhlman, R. L. In Topics in Organometallic Chemistry;
Guan, Z., Ed.; Springer: Berlin, 2009; Vol. 26, pp 65-104.
(15) (a) Wei, J.; Zhang, W.; Sita, L. Angew. Chem., Int. Ed. 2010, 49,
1768–1772. (b) Sita, L. R. Angew. Chem., Int. Ed. 2009, 48, 2464–2472.
(c) Zhang, W.; Sita, L. R. J. Am. Chem. Soc. 2008, 130, 442-443.
(d) Zhang, W.; Wei, J.; Sita, L. R. Macromolecules 2008, 41, 7829–7833.
(16) (a) Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1995,
497, 55–59. (b) Bochmann, M.; Lancaster, S. J. Angew. Chem., Int. Ed.
1994, 33, 1634–1637.
1
presence of oligomeric materials (Figure S-11). H NMR (400 MHz,
CDCl3): δ 1.16 (m, CH3(CH2)6CH3, 13H), 0.87 (m, CH3(CH2)6CH3,
1H). MALDI-TOF and FD-MS analyses of these oligomeric minor
products were inconclusive. The polymer was analyzed by GPC and
found to have Mn = 67 500, Mw= 172 500, PDI = 2.56 (Figure S-12). 1H
NMR analysis of the polymer (Figure S-13) showed very little
detectable unsaturation.
Chain Growth Using Al(oct)3. The reaction was carried out as
described in the general procedures with 4.39 ꢀ 10-6 M [rac-(EBI)-
Zr(μ-Me)2AlMe2þ][B(C6F5)4-], 20 psig ethylene, 0.439 M in Al.
Ethylene uptake plots deviated slightly from linearity after ethylene
saturation. The rate estimated from the ethylene-saturated region was
about 0.18 SCC/s. Hydrolysis of the resulting quenched product
mixture was carried out by pouring the reaction mixture into 300 mL
of 20% HCl (aq) with vigorous stirring. Polymeric product (176 mg)
immediately precipitated and was collected by filtering the biphasic
solution. The organic layer of the filtrate was then separated, and the
aqueous layer washed with toluene (2 ꢀ 25 mL). The combined organic
layers were washed with 1.0 M NaHCO3 (aq) and dried over Na2SO4.
1H NMR analysis of the toluene solution of product indicated only the
presence of octane (Figure S-14). 1H NMR (400 MHz, CDCl3): δ 1.26
(b, CH3(CH2)6CH3, 13H), 0.88 (t, CH3(CH2)6CH3, JH-H = 5.4 Hz,
1H). GPC of the resuting polymer (Figure S-15) showed it to have
Mw = 127 500, Mn = 56 000, and PDI = 2.26. 1H NMR of the polymer
(Figure S-16) showed it to have a small but detectable amount of vinyl
content.
’ ASSOCIATED CONTENT
S
Supporting Information. Tables S-1 through S-5, Figures
b
S-1 through S-16, derivation of the rate law for an alternative
mechanism for carboalumination, and details of the polymer
reactor design are available. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
jrn11@columbia.edu
5272
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273

Journal of the American Chemical Society
ARTICLE
(17) Petros, R. A.; Norton, J. R. Organometallics 2004, 23, 5105–
5107.
(38) Partial noncompetitive inhibition can be distinguished kineti-
cally from competitive inhibition by increasing the concentration of
inhibitor. This is impractical in our system due to the unfavorable
monomer/dimer equilibrium of AlMe3.
(39) Sydora, O. L.; Kilyanek, S. M.; Jordan, R. F. J. Am. Chem. Soc.
2007, 129, 12952-12953.
(40) (a) Landis, C. R.; Rosaaen, K. A.; Uddin, J. J. Am. Chem. Soc.
2002, 124, 12062–12063. (b) Dahlmann, M.; Erker, G.; Bergander, K.
J. Am. Chem. Soc. 2000, 122, 7986–7998.
(41) (a) Gately, D. A.; Norton, J. R.; Goodson, P. A. J. Am. Chem. Soc.
1995, 117, 986–996. (b) Grossman, R. B. Ph.D. Thesis, Massachusets
Institute of Technology, 1992.
(18) (a) Lieber, S.; Brintzinger, H. H. Macromolecules 2000, 33,
9192–9199. (b) Pedeutour, J.-N.; Radhakrishnan, K.; Cramail, H.;
Deffieux, A. Macromol. Rapid Commun. 2001, 22, 1095–1123. (c)
Pedeutour, J.-N.; Radhakrishnan, K.; Cramail, H.; Deffieux, A. J. Mol.
Catal. A: Chem. 2002, 185, 119–125. (d) G€otz, C.; Rau, A.; Luft, G.
Macromol. Mater. Eng. 2002, 287, 16–22. (e) G€otz, C.; Rau, A.; Luft, G.
J. Mol. Catal. A: Chem. 2002, 184, 95–110. (f) G€otz, C.; Rau, A.; Luft, G.
Macromol. Symp. 2002, 178, 93–107. (g) Bryliakov, K. P.; Semikolenova,
N. V.; Yudaev, D. V.; Zakharov, V. A.; Brintzinger, H.-H.; Ystenes, M.;
Rytter, E.; Talsi, E. P. J. Organomet. Chem. 2003, 683, 92–102. (h)
Kempe, R. Chem.-Eur. J. 2007, 13, 2764–2773. (i) Bochmann, M.
Organometallics 2010, 29, 4711–4740.
(19) Bolton, P. D.; Clot, E.; Cowley, A. R.; Mountford, P. J. Am.
Chem. Soc. 2006, 128, 15005–15018.
(20) Shaughnessy, K. H.; Waymouth, R. M. Organometallics 1998,
17, 5728–5745.
(21) Lee, C. H.; Lee, S. J.; Park, J. W.; Kim, K. H.; Lee, B. Y.; Oh, J. S.
J. Mol. Catal. A: Chem. 1998, 132, 231–239.
(22) Chen, E. Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391–1434.
(23) (a) Song, F. Q.; Cannon, R. D.; Lancaster, S. J.; Bochmann, M.
J. Mol. Catal. A: Chem. 2004, 218, 21–28. (b) Chien, J. C. W.; Tsai, W.
M.; Rausch, M. D. J. Am. Chem. Soc. 1991, 113, 8570–8571. (c) Chen, E.
Y.-X.; Marks, T. J. Chem. Rev. 2000, 100, 1391–1434.
(24) Petros, R. A.; Camara, J. M.; Norton, J. R. J. Organomet. Chem.
2007, 692, 4768–4773.
(25) Excess AlMe3 was added so that the heterobimetallic 3 was the
only Zr species observable in the solution 1H NMR. Typically, 5-10 mL
were sufficient.
(26) Cerny, Z.; Fusek, J.; Kriz, O.; Hermanek, S.; Solc, M.; Casensky,
B. J. Organomet. Chem. 1990, 386, 157–165.
(27) Bochman and Lancaster (ref 16) have said that alkylaluminum
adducts like 3 “are coordinatively saturated and lack a vacant orbital
suitable for binding the olefinic substrate”. While their coordinative
saturation can be questioned (note the vacant nonbonding orbital in
Figure 4 of ref 19), there seems to be general agreement on their inability
to bind olefins.
(28) Cornish-Bowden, A. Fundamentals of enzyme kinetics; Brook-
field: Portland, London, 1995.
(29) The standard deviations given for each value of K take into
account the covariance between the least-squares intercept and the least-
squares slope.
(30) The uncertainties given in Table 1 result from a generous
estimate (A. D. Bain, McMaster, 2003) of the covariance of koff with
other parameters fitted by CiFIT (ref 31). The uncertainties previously
reported (ref 17) came directly from CiFIT and did not include
covariance.
(31) Bain, A. D.; Cramer, J. A. J. Magn. Reson. 1996, 118, 21–27.
(32) (a) Konobeev, B. I.; Lyapin, V. V. Khimicheskaya Promyshlen-
nost (Moscow, Russian Federation) 1967, 43, 114–16. (b) Jolley, J. E.;
Hildebrand, J. H. J. Am. Chem. Soc. 1958, 80, 1050–1054. (c) Coan, C.
R.; King, A. D. J. Chromatogr. 1969, 44, 429–436.
(33) (a) Smith, M. B. J. Organomet. Chem. 1970, 22, 273–281. (b)
Smith, M. B. J. Organomet. Chem. 1972, 46, 31–49. (c) Smith, M. B. J.
Organomet. Chem. 1972, 46, 211–217. (d) Smith, M. B. J. Organomet.
Chem. 1974, 70, 13–33. (e) Smith, M. B. J. Phys. Chem. 1967, 71, 364–
370.
(34) The flow controller and flow meter are only accurate at flow
rates below 300 SCC/min.
(35) Talsi, E. P.; Eilertsen, J. L.; Ystenes, M.; Rytter, E. J. Organomet.
Chem. 2003, 677, 10–14.
(36) Two small additional signals, at -0.24 and -0.57 (ratio
approximately 2:1), appear when Me3Al is first added to [Cp2ZrMe]
[B(C6F5)4]; they disappear when the AlMe3 exceeds 1 equiv. Their
assignment (perhaps a 2:1 Zr/Al adduct?) is unclear.
(37) Segel, I. H. Enzyme kinetics: behavior and analysis of rapid
equilibrium and steady state enzyme systems; Wiley: New York, 1975.
5273
dx.doi.org/10.1021/ja104032w |J. Am. Chem. Soc. 2011, 133, 5263–5273